- Email: [email protected]
Corticosteroid Therapy in the Age of Infliximab: Acute and 1-Year Outcomes in Newly Diagnosed Children With Crohn’s Disease
CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2006;4:1124 –1129
Corticosteroid Therapy in the Age of Infliximab: Acute and 1-Year Outcomes in Newly Diagnosed Children With Crohn’s Disease JAMES MARKOWITZ,* JEFFREY HYAMS,‡ DAVID MACK,§ NEAL LELEIKO,¶ JONATHAN EVANS,储 SUBRA KUGATHASAN,# MARIAN PFEFFERKORN,** ADAM MEZOFF,‡‡ JOEL ROSH,§§ VASUNDHARA TOLIA,储 储 ANTHONY OTLEY,¶¶ ANNE GRIFFITHS,## M. SUSAN MOYER,*** MARIA OLIVA–HEMKER,‡‡‡ ROBERT WYLLIE,§§§ ROBERT ROTHBAUM,储 储 储 ATHOS BOUSVAROS,¶¶¶ J. FERNANDO DEL ROSARIO,### SANDRA HALE,‡ and TRUDY LERER,‡ FOR THE PEDIATRIC IBD COLLABORATIVE RESEARCH GROUP *North Shore–Long Island Jewish Health System, New Hyde Park, New York; ‡Connecticut Children’s Medical Center, Hartford, Connecticut; §Children’s Hospital of Eastern Ontario, Ottawa, Ontario, Canada; ¶Hasbro Children’s Hospital, Providence, Rhode Island; 储Nemours Children’s Clinic, Jacksonville, Florida; #Medical College of Wisconsin, Milwaukee, Wisconsin; **Riley Hospital for Children, Indianapolis, Indiana; ‡‡Children’s Medical Center, Dayton, Ohio; §§Morristown Memorial Hospital, Morristown, New Jersey; 储 储Children’s Hospital of Michigan, Detroit, Michigan; ¶¶IWK Health Centre, Halifax, Nova Scotia, Canada; ##Hospital for Sick Children, Toronto, Ontario, Canada; ***Children’s Hospital, Cincinnati, Ohio; ‡‡‡Johns Hopkins Hospital, Baltimore, Maryland; §§§Cleveland Clinic, Cleveland, Ohio; 储 储 储St. Louis Children’s Hospital, St. Louis, Missouri; ¶¶¶Children’s Hospital, Boston, Massachusetts; ###AI DuPont Hospital for Children, Wilmington, Delaware
See editorial on page 1094. Background & Aims: The aim of this study was to describe 3-month and 1-year outcomes of children with Crohn’s disease (CD) treated with corticosteroids within 30 days of diagnosis, with particular emphasis on the influence of infliximab on these outcomes. We also aimed to determine whether there are clinical or laboratory characteristics associated with corticosteroid therapy outcomes. Methods: Data from 109 children were drawn from a multicenter observational registry that was started in 2002. Clinical characteristics and data on corticosteroid and other therapies were recorded prospectively. Corticosteroid therapy outcomes at 3 months were defined as complete acute response, partial response, or corticosteroid resistance. At 1 year, corticosteroid responsiveness, dependence, and surgical rates were determined. Infliximab’s influence on short- and long-term outcomes also was investigated. Results: At 3 months, 65 of 109 (60%) patients had a complete acute response to corticosteroids, 26 (24%) had a partial response, and 18 (17%) were corticosteroid resistant. At 1 year, 61% were corticosteroid responsive, 31% were corticosteroid dependent, and 8% required surgery. Irrespective of the duration of corticosteroid treatment, 16 of 24 of corticosteroid-dependent/resistant patients rapidly discontinued corticosteroids after starting infliximab. No clinical or laboratory characteristics at diagnosis predicted short-term outcome. Growth impairment at diagnosis increased risk for corticosteroid dependence or surgery at 1 year. Conclusions: At 3 months, 84% of children had a complete or partial response to corticosteroids. However, despite concomitant immunomodulators, at 1 year 31% were corticosteroid dependent and 8% required surgery. Infliximab improves outcomes of corticosteroid-dependent/resistant patients because the duration of corticosteroid use can be controlled by initiating treatment with infliximab.
C
orticosteroids have been the mainstay of treatment for children with moderate-severe Crohn’s disease (CD). Although the early use of immunomodulatory drugs such as 6-mercaptopurine (6-MP), azathioprine, and methotrexate has
increased in an apparent attempt to decrease the need for corticosteroids, large numbers of children continue to require long-term treatment with these latter agents.1 Corticosteroids induced a complete remission of CD within 3 months in 58% of an adult cohort, but after 1 year only 32% were corticosteroid free, 28% were corticosteroid dependent, and 38% required surgery.2 Although no comparable population-based data in children with CD exist, a controlled trial in newly diagnosed children with moderate-severe CD treated only with prednisone resulted in prolonged remission in 50%, corticosteroid dependence in 39%, and corticosteroid resistance in 11%.3 These data, however, predate the commercial availability of infliximab. Small retrospective case series have documented marked reductions in concomitant corticosteroid therapy in children with CD.4,5 It is possible that the introduction of this potent biological agent to the therapeutic armamentarium may have changed the approach of many physicians to the therapy of children with CD. We have therefore used a prospective, multicenter, observational database of newly diagnosed children and adolescents with inflammatory bowel disease (IBD) to determine the course of CD in children who received corticosteroids within 30 days of initial diagnosis. Specifically, we (1) sought to determine both the short- and long-term responses of these children to corticosteroid therapy, (2) evaluated whether there are clinical or laboratory characteristics that are associated with a child’s response to corticosteroids, and (3) identified the frequency with which these children received treatment with infliximab and the effect of infliximab treatment on corticosteroid outcome during the first year after diagnosis.
Abbreviations used in this paper: CD, Crohn’s disease; IBD, inflammatory bowel disease; 6-MP, 6-mercaptopurine; PCDAI, Pediatric Crohn’s Disease Activity Index; Q, quarter. © 2006 by the American Gastroenterological Association (AGA) Institute 1542-3565/06/$32.00 doi:10.1016/j.cgh.2006.05.011
September 2006
Methods Patients for analysis were identified from a prospective database (Pediatric IBD Registry) compiled by members of the Pediatric Inflammatory Bowel Disease Collaborative Research Group. The Registry project has been reviewed and approved by the institutional review boards of each of the participating centers. The Registry was initiated in January 2002 by 18 US and Canadian pediatric gastroenterology centers with active IBD practices (Appendix 1; see supplemental material online at www.cghjournal.org). The data lock for the current investigation occurred in November of 2004. Investigators at each of the collaborating centers prospectively enrolled newly diagnosed children with IBD who had not yet reached their 16th birthday at the time of IBD diagnosis. Written informed consent was obtained from all patients’ parents or legal guardians, and written assent was obtained from all children older than 9 years of age. Subjects’ clinical and demographic characteristics, including type and extent of IBD, disease activity assessment (determined by Physician Global Assessment and Pediatric Crohn’s Disease Activity Index [PCDAI]), treatment, complications, and need for hospitalization or surgery were recorded at the time of initial diagnosis, 30 days after diagnosis, and quarterly thereafter. Routine laboratory data also were recorded prospectively. All children were managed according to the dictates of their physicians, not by standardized protocols. However, indications for the use of corticosteroids were similar among all treatment centers. Virtually all patients received corticosteroids as treatment for moderate-severe CD activity. In most patients corticosteroid therapy was initiated as oral prednisone or intravenous methylprednisolone at a dose of 1–2 mg/kg/day. For the current study, the database was searched for children with CD who were followed-up for at least 1 year and who received oral or parenteral corticosteroids within the first 30 days after diagnosis. Data on various therapies, including the use of corticosteroids; immunomodulatory agents including 6-MP, azathioprine, and methotrexate; infliximab; mesalamine; enteral nutrition; and surgery during the year after CD diagnosis were compiled prospectively.
Outcome Measures Short-term response. Acute responses to corticosteroid therapy were determined at 3 months after diagnosis. Children were considered to have a complete acute response if corticosteroids were stopped successfully within 3 months without requiring infliximab or surgery. Patients were considered corticosteroid resistant if they required either infliximab or surgery within 3 months after initiation of corticosteroid therapy. Those who remained on corticosteroids throughout the 3-month period of observation without requiring infliximab or surgery were considered to have a partial clinical response. Long-term response. Long-term response to corticosteroid therapy was assessed at 1 year. Patients were considered to be corticosteroid responsive if they discontinued corticosteroids within 3– 6 months of initiating therapy, and stayed off corticosteroids for the remainder of the first year. Corticosteroid-dependent patients received continuous corticosteroids for
CORTICOSTEROIDS IN PEDIATRIC CROHN’S DISEASE
1125
more than 6 months, or initially required corticosteroids for 3 months or less but required additional corticosteroids again in the first year to control symptoms. Corticosteroid-refractory patients required intra-abdominal surgery at any time during the year of observation. To determine the influence of infliximab on corticosteroid outcome, an additional assessment was made. Patients were considered to have a prolonged clinical response to corticosteroid therapy if they were weaned off corticosteroids by 3– 6 months without surgery or infliximab, and remained off corticosteroids at 1 year without additional corticosteroids or infliximab. Patients who required continuous corticosteroids beyond 6 months, repeated courses of corticosteroids throughout the first year, or who remained off corticosteroids associated with the administration of infliximab were considered to be on chronic corticosteroid/infliximab therapy.
Statistical Analysis Continuous variables, expressed as mean ⫾ SD, were compared between groups using t tests and analysis of variance. Categoric variables were compared using the Fisher exact test. All tests were 2-tailed and used a significance level of .05.
Patients At the time that the database was locked for analysis there were 403 children with CD enrolled in the Registry, of whom 186 had been followed-up for at least 1 year. Of these 186 children, 56 had mild disease and 130 had moderate-severe disease activity by Physician Global Assessment at the time of presentation. None of the 56 children with mild CD received corticosteroid treatment within the first 30 days after diagnosis. A total of 109 of the 130 children with moderate-severe CD met inclusion criteria for the present analysis, having received oral or parenteral corticosteroids within the first 30 days after disease diagnosis. These 109 constitute the study population. Their mean age at diagnosis was 11.8 ⫾ 2.8 years. Fifty-nine percent were male and 89% were Caucasian. The mean PCDAI score at diagnosis for this group was 35 ⫾ 16.
Results Overall Corticosteroid Use The number of patients receiving corticosteroid therapy continuously from treatment inception is shown in Figure 1. Although all patients, by definition, required corticosteroids in the first 30 days after CD diagnosis, there were progressively fewer children continuously receiving corticosteroids over the course of the first year of treatment.
Short-term Response to Corticosteroid Therapy By 3 months after diagnosis, 65 of 109 (60%) patients were weaned off corticosteroids completely without requiring infliximab or surgery (complete acute response) and an additional 26 patients (24%) had a partial response. Eighteen (17%) were resistant to corticosteroid therapy, including 2 who required surgery within 30 days of diagnosis and another 2 who had surgery by 3 months despite concomitant treatment with an immunomodulator (6-MP, azathioprine, or methotrexate). Ten of the resistant patients ultimately discontinued corticosteroids by 3 months after treatment with infliximab, so that a total of 75 of 109
1126
MARKOWITZ ET AL
CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 4, No. 9
(69%) patients discontinued corticosteroids by 3 months without surgery. There were no significant differences in the baseline PCDAI scores for the 3 subgroups of corticosteroid response (complete acute response PCDAI, 36.6 ⫾ 16.4; partial-response PCDAI, 30.9 ⫾ 14.2; and corticosteroid-resistant PCDAI, 40.0 ⫾ 18.9). Thirty-three of the 65 (51%) complete acute responders received concomitant 6-MP, azathioprine, or methotrexate within 3 months after diagnosis, whereas the other 32 were weaned to mesalamine medications or treated with enteral nutrition. The baseline PCDAI scores were similar in those acutely responding patients ultimately weaned to immunomodulators (38.1 ⫾ 17.1) and those weaned to mesalamine or enteral therapy (34.9 ⫾ 15.9). Compared with the group of complete acute responders, a significantly greater proportion of partially responsive and resistant patients (34/45 [76%]) received immunomodulatory drugs in the first 3 months after diagnosis (P ⫽ .01).
Long-term Response to Corticosteroid Therapy Long-term response is based on observation of only 102 patients because 7 lacked adequate data for analysis in at least 1 quarter of follow-up evaluation beyond 90 days. Eighty-two of 102 patients (80%) did not receive corticosteroids in the fourth quarter of observation. However, only 62 of 102 (61%) could be considered to have a prolonged corticosteroid response at 1 year, having discontinued corticosteroids by 3– 6 months without surgery. Forty-five of the 62 (73%) were maintained on 6-MP, azathioprine, or methotrexate after the corticosteroids were discontinued. The remainder received a mesalamine preparation and/or enteral feeding. Thirty-two (31%) patients had a long-term course characterized by corticosteroid dependence, although 12 ultimately were weaned off corticosteroids before the end of the year. Eleven of these 12 received concomitant 6-MP, azathioprine, or methotrexate. Although 17 of the remaining 20 patients also received immunomodulators, they required either continuous (n ⫽ 10) or repeated (n ⫽ 10) courses of corticosteroids throughout the year. Eight patients (8%) required surgery during the course of the year. In addition to the 4 patients who had surgery by 3 months, an additional patient had surgery in quarter 2, 2 in quarter 3, and 1 in quarter 4. Figure 2 summarizes the 1-year outcome of the study patients as a function of their 3-month response. Nearly three quarters of the acute corticosteroid responders experienced a prolonged response, compared with only 38% of the partial
Figure 1.
Percent of the population on continuous corticosteroid therapy over time.
Figure 2.
One-year response to corticosteroid (CS) treatment as a function of 3-month response. CS, corticosteroid. , Surgery; □, CS dependent; o, prolonged response.
responders and 53% of the acutely resistant patients. By contrast, only 5% of the complete acute responders and 4% of the partial responders required surgery compared with 24% of the acutely resistant patients. These differences are highly statistically significant (P ⫽ .0013).
Concomitant Therapies Among those patients who were weaned off corticosteroids successfully at some point during the year of observation, the therapies associated with steroid discontinuation included immunomodulatory agents alone (46%), immunomodulatory agents plus infliximab (19%), infliximab alone (1%), and surgery (3%). An additional 30% were weaned from corticosteroids with concomitant mesalamine medications or enteral feeds. Overall, 88 of 109 (81%) patients received immunomodulatory agents and 30 of 109 (28%) received infliximab at some point during the year of observation.
Infliximab Among the 30 children who received infliximab in the first year after diagnosis of CD were 15 of the 18 patients considered resistant to corticosteroids at 3 months. Another 3 patients started infliximab in quarter 2, 7 in quarter 3, and 5 in quarter 4. Six children were given infliximab as an alternative to another course of corticosteroids after they experienced an exacerbation of CD activity at a time after their initial corticosteroid course had been discontinued. Therefore, 24 children were given infliximab because of corticosteroid resistance or dependence. Irrespective of when in the course of observation infliximab was started, 16 of 24 (67%) children discontinued all corticosteroids within the same quarter without requiring surgery. Another 2 of these children only initiated infliximab in the fourth quarter of observation. Two children failed infliximab and required surgery (1 in quarter 1, 1 in quarter 4). Eighteen children initiated infliximab in the first 6 months after diagnosis. Nine (50%) were maintained on chronic infliximab, receiving 5 or more infliximab infusions by the end of the
September 2006
CORTICOSTEROIDS IN PEDIATRIC CROHN’S DISEASE
year. The other 9 were given between 1 and 4 infusions as an induction regimen, without subsequent infliximab maintenance. The duration of follow-up evaluation was too short to determine whether the 12 who received their initial infliximab dose between 7 and 12 months were, in fact, on maintenance infliximab at year’s end.
Discussion
Overall Outcome Figure 3 summarizes both the acute and 1-year outcomes of our study population. Despite the fact that 69% of the study population received concomitant immunomodulator therapy within 3 months of the initial diagnosis of CD, only 60% of the patients had a complete acute response at 90 days. Although 80% were off corticosteroids in the fourth quarter of observation, only 61% could be considered corticosteroid responsive at 1 year, whereas 31% were corticosteroid dependent and 8% required surgery. In addition, a prolonged clinical response without the use of infliximab was seen in only 46% of patients.
Relationship Between Baseline Patient Characteristics and Corticosteroid Response An extensive list of baseline demographic and clinical characteristics was evaluated to determine whether any were associated with a particular acute response to corticosteroids. No statistically significant differences were found among the 3 acute-response groups for age at diagnosis, sex, Tanner stage, disease activity as assessed by either Physician Global Assessment or PCDAI, site of CD involvement, frequency of extraintestinal manifestations, presence of concomitant medical conditions, or family history of IBD. In addition, the percentages of each acute-response group manifesting moderate or severe degrees of weight loss or growth retardation at diagnosis also were similar. There also were no significant differences in laboratory measures at diagnosis including hemoglobin level, white blood cell count, platelet count, erythrocyte sedimentation rate, and serum albumin level. However, differences in some of these baseline parameters were noted among the different 1-year response groups. The most significant difference was the frequency of baseline growth impairment (height percentile at least 1 major growth channel below pre-illness percentile) and growth failure (height percentile at least 2 major growth channels below pre-illness percentile). Sixty-seven percent of both the corticosteroid-dependent and surgical patients presented with growth impairment, compared with only 18% of those who had a prolonged 1-year response (P ⬍ .001). Growth failure at baseline was seen in 50% of those requiring surgery, 20% of the corticosteroid-dependent patients, and 10% of the corticosteroid-responsive patients (P ⫽ .048). However, the proportions of each 1-year response group with moderate and severe weight loss at diagnosis were similar. Severe CD activity at baseline as judged by Physician Global Assessment was noted in 62.5% of the patients ultimately requiring surgery, as compared with only 12.5% of the corticosteroid-dependent patients and 31% of the prolonged responders (P ⫽ .020). The use of immunomodulators and hospitalization by quarter 1 also was more common in the corticosteroid-dependent and surgery groups than in the corticosteroid-responsive group (P ⫽ .046 and .001, respectively).
1127
Population-based studies in adult patients with CD have shown that corticosteroid therapy is associated with a high rate of short-term response. In a study from Olmsted County, Minnesota, immediate outcomes 30 days after starting corticosteroids included complete remission in 58%, partial remission in 26%, and no response in 12%.2 Similar data (48% complete remission, 32% partial remission, 20% no response) have been reported from Copenhagen County, Denmark.6 In both studies, however, the long-term results of corticosteroid therapy were disappointing. One year after the initiation of corticosteroids, long-term response was seen in only 32% of the Olmsted County cohort, whereas 28% of patients were corticosteroid dependent and 38% required surgery.2 In the Danish study, among the patients showing either complete or partial remission at 30 days, only 55% remained in prolonged remission after treatment was discontinued, and 45% relapsed after corticosteroid weaning or could not be withdrawn without increased CD activity.6 Overall 1-year outcomes to a course of corticosteroids in this group were characterized as prolonged response in 44%, corticosteroid dependency in 36%, and corticosteroid resistance in 20%.6 Comparable data in a pediatric population have not been reported, although reports have suggested high risks of corticosteroid dependence or resistance in children.7,8 Although not designed to address children’s responses directly to corticosteroid therapy, a multicenter prospective trial used prednisone as the control arm of a study assessing the response of newly diagnosed children with moderate-severe CD to treatment with prednisone and 6-MP.3 In this study, corticosteroid therapy was standardized across all centers, with weaning or eventual discontinuation of prednisone determined by change in a patient’s disease activity score over time. When the outcomes of the control patients who had received prednisone plus a placebo 6-MP pill were analyzed, short-term clinical response to prednisone at 30 days, indicated by inactive disease activity scores, was seen in 79% of children. However, because prednisone was weaned or discontinued, relapses were common. By 18 months after corticosteroids were initiated, a prolonged response (remission of CD without need for additional prednisone, other medication, or surgery) was seen in only 50% of the group, whereas 39% were prednisone dependent and 11% were prednisone resistant.3
Figure 3.
Summary of acute and 1-year outcomes.
1128
MARKOWITZ ET AL
The data from the current study confirm that pediatric gastroenterologists continue to use corticosteroids as a first-line therapy for children with newly diagnosed moderate-severe CD. In fact, 109 of 130 (84%) of the children with moderate-severe CD activity at baseline who were enrolled in the Registry at the time of this study received corticosteroid therapy within the first 30 days after diagnosis. Our data reveal that short-term responses to corticosteroids in an uncontrolled clinical setting are very similar to those previously identified in both adult populations and the controlled trial described earlier.2,3,6 Successful weaning of corticosteroids by 3 months without the need for infliximab or surgery was seen in 60% of all patients. Still, at 3 months, 24% were corticosteroid dependent and 17% required infliximab or surgery. It is of some interest that the increased use of concomitant immunomodulatory medications did not appear to improve the short-term response to corticosteroids. Many more children (69% by quarter 1) in the present study received immunomodulators than did the adult patients (3%) in the Olmsted County study.2 Presumably the onset of action of 6-MP and azathioprine is too delayed to influence the 3-month response rate to corticosteroids significantly. This supposition is supported by the results of the placebo-controlled multicenter pediatric 6-MP trial, which revealed no differences in the rates of response or remission at 3 months between the group receiving only prednisone and the group receiving both prednisone and 6-MP.3 Possibly more important, however, is that the long-term response to corticosteroids appears to be better than that reported previously because 61% of patients showed corticosteroid responsiveness, 31% showed dependence, and only 8% required surgery. The previously cited multicenter pediatric 6-MP trial clearly showed that prolonged remission without the need for corticosteroids is common when 6-MP is started early in the course of a child with moderate-severe CD.3 It seems likely, therefore, that the widespread use of immunomodulators (81% by year’s end) is an important reason for the improved long-term response of our patients. Progressively fewer children received corticosteroids over the course of the year. In fact, only 10 patients (9%) were treated continuously with corticosteroids throughout the year of observation. However, despite the ongoing use of immunomodulators, in many cases corticosteroids could be discontinued only after the initiation of infliximab therapy. Among the corticosteroid-dependent or corticosteroidresistant children, 67% of those treated with infliximab were weaned off corticosteroids successfully only after starting infliximab. The 8% surgical rate at 1 year in the present study is distinctly lower than the 38% rate identified in the data derived from an adult population of patients with CD drawn from Olmsted County.2 A number of factors likely explain this difference. As opposed to the current study in newly diagnosed children, the duration of CD at the initiation of corticosteroids in the Olmsted County study ranged from 0 to 18.4 years. It therefore is likely that the Olmsted County population included a greater number of patients with long-standing fibrostenosing disease that might be less responsive to long-term medical intervention. The marked discrepancy in rates of immunomodulator use (81% vs 3%) also is significant. In addition, the Olmsted County study predates the introduction of infliximab into clinical practice. Finally, because our patients were newly diagnosed children, there also might be a bias among
CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 4, No. 9
pediatric gastroenterologists and parents toward continued medical therapy and the avoidance of surgery. Despite evaluation of a wide spectrum of demographic, clinical, and laboratory characteristics, we could not identify factors at diagnosis that were associated with a patient’s short-term response to corticosteroids. By contrast, our data show that children who have established growth impairment at the time of CD diagnosis are at higher risk for corticosteroid dependence and surgery during the first year. This represents a particularly difficult problem given the well-recognized growth-suppressive effect of corticosteroid therapy itself. Prolonged corticosteroid therapy likely further will aggravate the growth problems in these children. Clearly, the newly diagnosed child with moderate-severe CD activity and growth impairment will require a different treatment strategy that will avoid the use of corticosteroids. It appears, however, that infliximab can change the outcome of corticosteroid therapy in children with moderate-severe CD. Corticosteroid dependence is not an inescapable outcome of treatment for many children. The corticosteroid-sparing effect of infliximab allows physicians to decide how long to use corticosteroids and provides them with a highly successful management tool to promote discontinuation of these agents. Because many of the children in this study did not begin infliximab until the final 6 months of observation, the duration of follow-up evaluation was too short to determine how many patients required chronic infliximab therapy. However, a recent Danish retrospective analysis of children with CD treated with infliximab showed that only 29% experienced a prolonged clinical response after an induction regimen without subsequent infusions, and that another 42% required maintenance therapy.9 If long-term infliximab is ultimately necessary to keep children with CD free of corticosteroids, studies will be needed to determine the safety and efficacy of maintenance infliximab in this population.
Supplementary Data Note: To access the supplementary material accompanying this article, visit the online version of Clinical Gastroenterology and Hepatology at www.cghjournal.org. References 1. Markowitz J, Grancher K, Kohn N, et al. Immunomodulatory therapy for pediatric inflammatory bowel disease: changing patterns of use, 1990 –2000. Am J Gastroenterol 2002;97:928 –932. 2. Faubion WA Jr, Loftus EV Jr, Harmsen WS, et al. The natural history of corticosteroid therapy for inflammatory bowel disease: a population-based study. Gastroenterology 2001;121:255–260. 3. Markowitz J, Grancher K, Kohn N, et al. A multicenter trial of 6-mercaptopurine and prednisone in children with newly diagnosed Crohn’s disease. Gastroenterology 2000;119:895–902. 4. Hyams JS, Markowitz J, Wyllie R. Use of infliximab in the treatment of Crohn’s disease in children and adolescents. J Pediatr 2000; 137:192–196. 5. Lamireau T, Cezard JP, Dabadie A, et al. Efficacy and tolerance of infliximab in children and adolescents with Crohn’s disease. Inflamm Bowel Dis 2004;10:745–750. 6. Munkholm P, Langholz E, Davidsen M, et al. Frequency of glucocorticoid resistance and dependency in Crohn’s disease. Gut 1994; 35:360 –362. 7. Escher JC, Taminiau JA, Nieuwenhuis EE, et al. Treatment of inflammatory bowel disease in childhood: best available evidence. Inflamm Bowel Dis 2003;9:34 –58.
September 2006
8. Beattie RM. Therapy of Crohn’s disease in childhood. Paediatr Drugs 2000;2:193–203. 9. Wewer V, Riis S, Vind I, et al. Infliximab dependency in a national cohort of children with Crohn’s disease. J Pediatr Gastroenterol Nutr 2006;42:40 – 45.
Address requests for reprints to: James Markowitz, MD, Division of Pediatric Gastroenterology, Schneider Children’s Hospital, 269-01 76th Avenue, New Hyde Park, New York 11040. e-mail: jmarkowi@ nshs.edu; fax: (718) 962-2908. Supported by grants from Centocor, Inc. (Malvern, PA), AstraZeneca (Wilmington, DE), Reach Out for Youth With Ileitis and Colitis (Melville, NY), and the collaborating institutions. The operational expenses for the Registry Data Center are supported by grants from Centocor (Horsham, PA), AstraZeneca (Wilmington, DE), and Reach Out for Youth with Ileitis
CORTICOSTEROIDS IN PEDIATRIC CROHN’S DISEASE
1129
and Colitis (Melville, NY), and from grants from the participating centers. Vasundhara Tolia has received grants, honoraria, and/or consultancy fees from Centocor. Drs Markowitz, Hyams, and Rosh have received honoraria and/or consultancy fees from Prometheus Labs. James Markowitz has received research support from and has served as a Consultant to Centocor. Jeffrey Hyams has received research support from Centocor, and has served as a Consultant to Centocor, Abbott, and ELAN. Subra Kugathasan participated as a Consultant (advisory panel in designing clinical trials) for Centocor and Abbott. Joel Rosh has received research grant support from Proctor and Gamble. The authors are deeply indebted to the following research coordinators whose efforts greatly facilitated the performance of this study: Patricia Davis, Kathy Grancher, Valerie Grant, Annette Langseder, Ruth Singleton, Anna Zholudev, George Kay, Gail Waltz, Kim Boyer, Shari Huffman, Cathy Williams, Rebecca Abood, Rosemary Nagy, Carol Rudman, Myrna Miller, Vivian Abadom, Janet Trotta, and Laura Defaveri.
Corticosteroid Therapy in the Age of Infliximab: Acute and 1-Year Outcomes in Newly Diagnosed Children With Crohn’s Disease JAMES MARKOWITZ,* JEFFREY HYAMS,‡ DAVID MACK,§ NEAL LELEIKO,¶ JONATHAN EVANS,储 SUBRA KUGATHASAN,# MARIAN PFEFFERKORN,** ADAM MEZOFF,‡‡ JOEL ROSH,§§ VASUNDHARA TOLIA,储 储 ANTHONY OTLEY,¶¶ ANNE GRIFFITHS,## M. SUSAN MOYER,*** MARIA OLIVA–HEMKER,‡‡‡ ROBERT WYLLIE,§§§ ROBERT ROTHBAUM,储 储 储 ATHOS BOUSVAROS,¶¶¶ J. FERNANDO DEL ROSARIO,### SANDRA HALE,‡ and TRUDY LERER,‡ FOR THE PEDIATRIC IBD COLLABORATIVE RESEARCH GROUP *North Shore–Long Island Jewish Health System, New Hyde Park, New York; ‡Connecticut Children’s Medical Center, Hartford, Connecticut; §Children’s Hospital of Eastern Ontario, Ottawa, Ontario, Canada; ¶Hasbro Children’s Hospital, Providence, Rhode Island; 储Nemours Children’s Clinic, Jacksonville, Florida; #Medical College of Wisconsin, Milwaukee, Wisconsin; **Riley Hospital for Children, Indianapolis, Indiana; ‡‡Children’s Medical Center, Dayton, Ohio; §§Morristown Memorial Hospital, Morristown, New Jersey; 储 储Children’s Hospital of Michigan, Detroit, Michigan; ¶¶IWK Health Centre, Halifax, Nova Scotia, Canada; ##Hospital for Sick Children, Toronto, Ontario, Canada; ***Children’s Hospital, Cincinnati, Ohio; ‡‡‡Johns Hopkins Hospital, Baltimore, Maryland; §§§Cleveland Clinic, Cleveland, Ohio; 储 储 储St. Louis Children’s Hospital, St. Louis, Missouri; ¶¶¶Children’s Hospital, Boston, Massachusetts; ###AI DuPont Hospital for Children, Wilmington, Delaware
See editorial on page 1094. Background & Aims: The aim of this study was to describe 3-month and 1-year outcomes of children with Crohn’s disease (CD) treated with corticosteroids within 30 days of diagnosis, with particular emphasis on the influence of infliximab on these outcomes. We also aimed to determine whether there are clinical or laboratory characteristics associated with corticosteroid therapy outcomes. Methods: Data from 109 children were drawn from a multicenter observational registry that was started in 2002. Clinical characteristics and data on corticosteroid and other therapies were recorded prospectively. Corticosteroid therapy outcomes at 3 months were defined as complete acute response, partial response, or corticosteroid resistance. At 1 year, corticosteroid responsiveness, dependence, and surgical rates were determined. Infliximab’s influence on short- and long-term outcomes also was investigated. Results: At 3 months, 65 of 109 (60%) patients had a complete acute response to corticosteroids, 26 (24%) had a partial response, and 18 (17%) were corticosteroid resistant. At 1 year, 61% were corticosteroid responsive, 31% were corticosteroid dependent, and 8% required surgery. Irrespective of the duration of corticosteroid treatment, 16 of 24 of corticosteroid-dependent/resistant patients rapidly discontinued corticosteroids after starting infliximab. No clinical or laboratory characteristics at diagnosis predicted short-term outcome. Growth impairment at diagnosis increased risk for corticosteroid dependence or surgery at 1 year. Conclusions: At 3 months, 84% of children had a complete or partial response to corticosteroids. However, despite concomitant immunomodulators, at 1 year 31% were corticosteroid dependent and 8% required surgery. Infliximab improves outcomes of corticosteroid-dependent/resistant patients because the duration of corticosteroid use can be controlled by initiating treatment with infliximab.
C
orticosteroids have been the mainstay of treatment for children with moderate-severe Crohn’s disease (CD). Although the early use of immunomodulatory drugs such as 6-mercaptopurine (6-MP), azathioprine, and methotrexate has
increased in an apparent attempt to decrease the need for corticosteroids, large numbers of children continue to require long-term treatment with these latter agents.1 Corticosteroids induced a complete remission of CD within 3 months in 58% of an adult cohort, but after 1 year only 32% were corticosteroid free, 28% were corticosteroid dependent, and 38% required surgery.2 Although no comparable population-based data in children with CD exist, a controlled trial in newly diagnosed children with moderate-severe CD treated only with prednisone resulted in prolonged remission in 50%, corticosteroid dependence in 39%, and corticosteroid resistance in 11%.3 These data, however, predate the commercial availability of infliximab. Small retrospective case series have documented marked reductions in concomitant corticosteroid therapy in children with CD.4,5 It is possible that the introduction of this potent biological agent to the therapeutic armamentarium may have changed the approach of many physicians to the therapy of children with CD. We have therefore used a prospective, multicenter, observational database of newly diagnosed children and adolescents with inflammatory bowel disease (IBD) to determine the course of CD in children who received corticosteroids within 30 days of initial diagnosis. Specifically, we (1) sought to determine both the short- and long-term responses of these children to corticosteroid therapy, (2) evaluated whether there are clinical or laboratory characteristics that are associated with a child’s response to corticosteroids, and (3) identified the frequency with which these children received treatment with infliximab and the effect of infliximab treatment on corticosteroid outcome during the first year after diagnosis.
Abbreviations used in this paper: CD, Crohn’s disease; IBD, inflammatory bowel disease; 6-MP, 6-mercaptopurine; PCDAI, Pediatric Crohn’s Disease Activity Index; Q, quarter. © 2006 by the American Gastroenterological Association (AGA) Institute 1542-3565/06/$32.00 doi:10.1016/j.cgh.2006.05.011
September 2006
Methods Patients for analysis were identified from a prospective database (Pediatric IBD Registry) compiled by members of the Pediatric Inflammatory Bowel Disease Collaborative Research Group. The Registry project has been reviewed and approved by the institutional review boards of each of the participating centers. The Registry was initiated in January 2002 by 18 US and Canadian pediatric gastroenterology centers with active IBD practices (Appendix 1; see supplemental material online at www.cghjournal.org). The data lock for the current investigation occurred in November of 2004. Investigators at each of the collaborating centers prospectively enrolled newly diagnosed children with IBD who had not yet reached their 16th birthday at the time of IBD diagnosis. Written informed consent was obtained from all patients’ parents or legal guardians, and written assent was obtained from all children older than 9 years of age. Subjects’ clinical and demographic characteristics, including type and extent of IBD, disease activity assessment (determined by Physician Global Assessment and Pediatric Crohn’s Disease Activity Index [PCDAI]), treatment, complications, and need for hospitalization or surgery were recorded at the time of initial diagnosis, 30 days after diagnosis, and quarterly thereafter. Routine laboratory data also were recorded prospectively. All children were managed according to the dictates of their physicians, not by standardized protocols. However, indications for the use of corticosteroids were similar among all treatment centers. Virtually all patients received corticosteroids as treatment for moderate-severe CD activity. In most patients corticosteroid therapy was initiated as oral prednisone or intravenous methylprednisolone at a dose of 1–2 mg/kg/day. For the current study, the database was searched for children with CD who were followed-up for at least 1 year and who received oral or parenteral corticosteroids within the first 30 days after diagnosis. Data on various therapies, including the use of corticosteroids; immunomodulatory agents including 6-MP, azathioprine, and methotrexate; infliximab; mesalamine; enteral nutrition; and surgery during the year after CD diagnosis were compiled prospectively.
Outcome Measures Short-term response. Acute responses to corticosteroid therapy were determined at 3 months after diagnosis. Children were considered to have a complete acute response if corticosteroids were stopped successfully within 3 months without requiring infliximab or surgery. Patients were considered corticosteroid resistant if they required either infliximab or surgery within 3 months after initiation of corticosteroid therapy. Those who remained on corticosteroids throughout the 3-month period of observation without requiring infliximab or surgery were considered to have a partial clinical response. Long-term response. Long-term response to corticosteroid therapy was assessed at 1 year. Patients were considered to be corticosteroid responsive if they discontinued corticosteroids within 3– 6 months of initiating therapy, and stayed off corticosteroids for the remainder of the first year. Corticosteroid-dependent patients received continuous corticosteroids for
CORTICOSTEROIDS IN PEDIATRIC CROHN’S DISEASE
1125
more than 6 months, or initially required corticosteroids for 3 months or less but required additional corticosteroids again in the first year to control symptoms. Corticosteroid-refractory patients required intra-abdominal surgery at any time during the year of observation. To determine the influence of infliximab on corticosteroid outcome, an additional assessment was made. Patients were considered to have a prolonged clinical response to corticosteroid therapy if they were weaned off corticosteroids by 3– 6 months without surgery or infliximab, and remained off corticosteroids at 1 year without additional corticosteroids or infliximab. Patients who required continuous corticosteroids beyond 6 months, repeated courses of corticosteroids throughout the first year, or who remained off corticosteroids associated with the administration of infliximab were considered to be on chronic corticosteroid/infliximab therapy.
Statistical Analysis Continuous variables, expressed as mean ⫾ SD, were compared between groups using t tests and analysis of variance. Categoric variables were compared using the Fisher exact test. All tests were 2-tailed and used a significance level of .05.
Patients At the time that the database was locked for analysis there were 403 children with CD enrolled in the Registry, of whom 186 had been followed-up for at least 1 year. Of these 186 children, 56 had mild disease and 130 had moderate-severe disease activity by Physician Global Assessment at the time of presentation. None of the 56 children with mild CD received corticosteroid treatment within the first 30 days after diagnosis. A total of 109 of the 130 children with moderate-severe CD met inclusion criteria for the present analysis, having received oral or parenteral corticosteroids within the first 30 days after disease diagnosis. These 109 constitute the study population. Their mean age at diagnosis was 11.8 ⫾ 2.8 years. Fifty-nine percent were male and 89% were Caucasian. The mean PCDAI score at diagnosis for this group was 35 ⫾ 16.
Results Overall Corticosteroid Use The number of patients receiving corticosteroid therapy continuously from treatment inception is shown in Figure 1. Although all patients, by definition, required corticosteroids in the first 30 days after CD diagnosis, there were progressively fewer children continuously receiving corticosteroids over the course of the first year of treatment.
Short-term Response to Corticosteroid Therapy By 3 months after diagnosis, 65 of 109 (60%) patients were weaned off corticosteroids completely without requiring infliximab or surgery (complete acute response) and an additional 26 patients (24%) had a partial response. Eighteen (17%) were resistant to corticosteroid therapy, including 2 who required surgery within 30 days of diagnosis and another 2 who had surgery by 3 months despite concomitant treatment with an immunomodulator (6-MP, azathioprine, or methotrexate). Ten of the resistant patients ultimately discontinued corticosteroids by 3 months after treatment with infliximab, so that a total of 75 of 109
1126
MARKOWITZ ET AL
CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 4, No. 9
(69%) patients discontinued corticosteroids by 3 months without surgery. There were no significant differences in the baseline PCDAI scores for the 3 subgroups of corticosteroid response (complete acute response PCDAI, 36.6 ⫾ 16.4; partial-response PCDAI, 30.9 ⫾ 14.2; and corticosteroid-resistant PCDAI, 40.0 ⫾ 18.9). Thirty-three of the 65 (51%) complete acute responders received concomitant 6-MP, azathioprine, or methotrexate within 3 months after diagnosis, whereas the other 32 were weaned to mesalamine medications or treated with enteral nutrition. The baseline PCDAI scores were similar in those acutely responding patients ultimately weaned to immunomodulators (38.1 ⫾ 17.1) and those weaned to mesalamine or enteral therapy (34.9 ⫾ 15.9). Compared with the group of complete acute responders, a significantly greater proportion of partially responsive and resistant patients (34/45 [76%]) received immunomodulatory drugs in the first 3 months after diagnosis (P ⫽ .01).
Long-term Response to Corticosteroid Therapy Long-term response is based on observation of only 102 patients because 7 lacked adequate data for analysis in at least 1 quarter of follow-up evaluation beyond 90 days. Eighty-two of 102 patients (80%) did not receive corticosteroids in the fourth quarter of observation. However, only 62 of 102 (61%) could be considered to have a prolonged corticosteroid response at 1 year, having discontinued corticosteroids by 3– 6 months without surgery. Forty-five of the 62 (73%) were maintained on 6-MP, azathioprine, or methotrexate after the corticosteroids were discontinued. The remainder received a mesalamine preparation and/or enteral feeding. Thirty-two (31%) patients had a long-term course characterized by corticosteroid dependence, although 12 ultimately were weaned off corticosteroids before the end of the year. Eleven of these 12 received concomitant 6-MP, azathioprine, or methotrexate. Although 17 of the remaining 20 patients also received immunomodulators, they required either continuous (n ⫽ 10) or repeated (n ⫽ 10) courses of corticosteroids throughout the year. Eight patients (8%) required surgery during the course of the year. In addition to the 4 patients who had surgery by 3 months, an additional patient had surgery in quarter 2, 2 in quarter 3, and 1 in quarter 4. Figure 2 summarizes the 1-year outcome of the study patients as a function of their 3-month response. Nearly three quarters of the acute corticosteroid responders experienced a prolonged response, compared with only 38% of the partial
Figure 1.
Percent of the population on continuous corticosteroid therapy over time.
Figure 2.
One-year response to corticosteroid (CS) treatment as a function of 3-month response. CS, corticosteroid. , Surgery; □, CS dependent; o, prolonged response.
responders and 53% of the acutely resistant patients. By contrast, only 5% of the complete acute responders and 4% of the partial responders required surgery compared with 24% of the acutely resistant patients. These differences are highly statistically significant (P ⫽ .0013).
Concomitant Therapies Among those patients who were weaned off corticosteroids successfully at some point during the year of observation, the therapies associated with steroid discontinuation included immunomodulatory agents alone (46%), immunomodulatory agents plus infliximab (19%), infliximab alone (1%), and surgery (3%). An additional 30% were weaned from corticosteroids with concomitant mesalamine medications or enteral feeds. Overall, 88 of 109 (81%) patients received immunomodulatory agents and 30 of 109 (28%) received infliximab at some point during the year of observation.
Infliximab Among the 30 children who received infliximab in the first year after diagnosis of CD were 15 of the 18 patients considered resistant to corticosteroids at 3 months. Another 3 patients started infliximab in quarter 2, 7 in quarter 3, and 5 in quarter 4. Six children were given infliximab as an alternative to another course of corticosteroids after they experienced an exacerbation of CD activity at a time after their initial corticosteroid course had been discontinued. Therefore, 24 children were given infliximab because of corticosteroid resistance or dependence. Irrespective of when in the course of observation infliximab was started, 16 of 24 (67%) children discontinued all corticosteroids within the same quarter without requiring surgery. Another 2 of these children only initiated infliximab in the fourth quarter of observation. Two children failed infliximab and required surgery (1 in quarter 1, 1 in quarter 4). Eighteen children initiated infliximab in the first 6 months after diagnosis. Nine (50%) were maintained on chronic infliximab, receiving 5 or more infliximab infusions by the end of the
September 2006
CORTICOSTEROIDS IN PEDIATRIC CROHN’S DISEASE
year. The other 9 were given between 1 and 4 infusions as an induction regimen, without subsequent infliximab maintenance. The duration of follow-up evaluation was too short to determine whether the 12 who received their initial infliximab dose between 7 and 12 months were, in fact, on maintenance infliximab at year’s end.
Discussion
Overall Outcome Figure 3 summarizes both the acute and 1-year outcomes of our study population. Despite the fact that 69% of the study population received concomitant immunomodulator therapy within 3 months of the initial diagnosis of CD, only 60% of the patients had a complete acute response at 90 days. Although 80% were off corticosteroids in the fourth quarter of observation, only 61% could be considered corticosteroid responsive at 1 year, whereas 31% were corticosteroid dependent and 8% required surgery. In addition, a prolonged clinical response without the use of infliximab was seen in only 46% of patients.
Relationship Between Baseline Patient Characteristics and Corticosteroid Response An extensive list of baseline demographic and clinical characteristics was evaluated to determine whether any were associated with a particular acute response to corticosteroids. No statistically significant differences were found among the 3 acute-response groups for age at diagnosis, sex, Tanner stage, disease activity as assessed by either Physician Global Assessment or PCDAI, site of CD involvement, frequency of extraintestinal manifestations, presence of concomitant medical conditions, or family history of IBD. In addition, the percentages of each acute-response group manifesting moderate or severe degrees of weight loss or growth retardation at diagnosis also were similar. There also were no significant differences in laboratory measures at diagnosis including hemoglobin level, white blood cell count, platelet count, erythrocyte sedimentation rate, and serum albumin level. However, differences in some of these baseline parameters were noted among the different 1-year response groups. The most significant difference was the frequency of baseline growth impairment (height percentile at least 1 major growth channel below pre-illness percentile) and growth failure (height percentile at least 2 major growth channels below pre-illness percentile). Sixty-seven percent of both the corticosteroid-dependent and surgical patients presented with growth impairment, compared with only 18% of those who had a prolonged 1-year response (P ⬍ .001). Growth failure at baseline was seen in 50% of those requiring surgery, 20% of the corticosteroid-dependent patients, and 10% of the corticosteroid-responsive patients (P ⫽ .048). However, the proportions of each 1-year response group with moderate and severe weight loss at diagnosis were similar. Severe CD activity at baseline as judged by Physician Global Assessment was noted in 62.5% of the patients ultimately requiring surgery, as compared with only 12.5% of the corticosteroid-dependent patients and 31% of the prolonged responders (P ⫽ .020). The use of immunomodulators and hospitalization by quarter 1 also was more common in the corticosteroid-dependent and surgery groups than in the corticosteroid-responsive group (P ⫽ .046 and .001, respectively).
1127
Population-based studies in adult patients with CD have shown that corticosteroid therapy is associated with a high rate of short-term response. In a study from Olmsted County, Minnesota, immediate outcomes 30 days after starting corticosteroids included complete remission in 58%, partial remission in 26%, and no response in 12%.2 Similar data (48% complete remission, 32% partial remission, 20% no response) have been reported from Copenhagen County, Denmark.6 In both studies, however, the long-term results of corticosteroid therapy were disappointing. One year after the initiation of corticosteroids, long-term response was seen in only 32% of the Olmsted County cohort, whereas 28% of patients were corticosteroid dependent and 38% required surgery.2 In the Danish study, among the patients showing either complete or partial remission at 30 days, only 55% remained in prolonged remission after treatment was discontinued, and 45% relapsed after corticosteroid weaning or could not be withdrawn without increased CD activity.6 Overall 1-year outcomes to a course of corticosteroids in this group were characterized as prolonged response in 44%, corticosteroid dependency in 36%, and corticosteroid resistance in 20%.6 Comparable data in a pediatric population have not been reported, although reports have suggested high risks of corticosteroid dependence or resistance in children.7,8 Although not designed to address children’s responses directly to corticosteroid therapy, a multicenter prospective trial used prednisone as the control arm of a study assessing the response of newly diagnosed children with moderate-severe CD to treatment with prednisone and 6-MP.3 In this study, corticosteroid therapy was standardized across all centers, with weaning or eventual discontinuation of prednisone determined by change in a patient’s disease activity score over time. When the outcomes of the control patients who had received prednisone plus a placebo 6-MP pill were analyzed, short-term clinical response to prednisone at 30 days, indicated by inactive disease activity scores, was seen in 79% of children. However, because prednisone was weaned or discontinued, relapses were common. By 18 months after corticosteroids were initiated, a prolonged response (remission of CD without need for additional prednisone, other medication, or surgery) was seen in only 50% of the group, whereas 39% were prednisone dependent and 11% were prednisone resistant.3
Figure 3.
Summary of acute and 1-year outcomes.
1128
MARKOWITZ ET AL
The data from the current study confirm that pediatric gastroenterologists continue to use corticosteroids as a first-line therapy for children with newly diagnosed moderate-severe CD. In fact, 109 of 130 (84%) of the children with moderate-severe CD activity at baseline who were enrolled in the Registry at the time of this study received corticosteroid therapy within the first 30 days after diagnosis. Our data reveal that short-term responses to corticosteroids in an uncontrolled clinical setting are very similar to those previously identified in both adult populations and the controlled trial described earlier.2,3,6 Successful weaning of corticosteroids by 3 months without the need for infliximab or surgery was seen in 60% of all patients. Still, at 3 months, 24% were corticosteroid dependent and 17% required infliximab or surgery. It is of some interest that the increased use of concomitant immunomodulatory medications did not appear to improve the short-term response to corticosteroids. Many more children (69% by quarter 1) in the present study received immunomodulators than did the adult patients (3%) in the Olmsted County study.2 Presumably the onset of action of 6-MP and azathioprine is too delayed to influence the 3-month response rate to corticosteroids significantly. This supposition is supported by the results of the placebo-controlled multicenter pediatric 6-MP trial, which revealed no differences in the rates of response or remission at 3 months between the group receiving only prednisone and the group receiving both prednisone and 6-MP.3 Possibly more important, however, is that the long-term response to corticosteroids appears to be better than that reported previously because 61% of patients showed corticosteroid responsiveness, 31% showed dependence, and only 8% required surgery. The previously cited multicenter pediatric 6-MP trial clearly showed that prolonged remission without the need for corticosteroids is common when 6-MP is started early in the course of a child with moderate-severe CD.3 It seems likely, therefore, that the widespread use of immunomodulators (81% by year’s end) is an important reason for the improved long-term response of our patients. Progressively fewer children received corticosteroids over the course of the year. In fact, only 10 patients (9%) were treated continuously with corticosteroids throughout the year of observation. However, despite the ongoing use of immunomodulators, in many cases corticosteroids could be discontinued only after the initiation of infliximab therapy. Among the corticosteroid-dependent or corticosteroidresistant children, 67% of those treated with infliximab were weaned off corticosteroids successfully only after starting infliximab. The 8% surgical rate at 1 year in the present study is distinctly lower than the 38% rate identified in the data derived from an adult population of patients with CD drawn from Olmsted County.2 A number of factors likely explain this difference. As opposed to the current study in newly diagnosed children, the duration of CD at the initiation of corticosteroids in the Olmsted County study ranged from 0 to 18.4 years. It therefore is likely that the Olmsted County population included a greater number of patients with long-standing fibrostenosing disease that might be less responsive to long-term medical intervention. The marked discrepancy in rates of immunomodulator use (81% vs 3%) also is significant. In addition, the Olmsted County study predates the introduction of infliximab into clinical practice. Finally, because our patients were newly diagnosed children, there also might be a bias among
CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 4, No. 9
pediatric gastroenterologists and parents toward continued medical therapy and the avoidance of surgery. Despite evaluation of a wide spectrum of demographic, clinical, and laboratory characteristics, we could not identify factors at diagnosis that were associated with a patient’s short-term response to corticosteroids. By contrast, our data show that children who have established growth impairment at the time of CD diagnosis are at higher risk for corticosteroid dependence and surgery during the first year. This represents a particularly difficult problem given the well-recognized growth-suppressive effect of corticosteroid therapy itself. Prolonged corticosteroid therapy likely further will aggravate the growth problems in these children. Clearly, the newly diagnosed child with moderate-severe CD activity and growth impairment will require a different treatment strategy that will avoid the use of corticosteroids. It appears, however, that infliximab can change the outcome of corticosteroid therapy in children with moderate-severe CD. Corticosteroid dependence is not an inescapable outcome of treatment for many children. The corticosteroid-sparing effect of infliximab allows physicians to decide how long to use corticosteroids and provides them with a highly successful management tool to promote discontinuation of these agents. Because many of the children in this study did not begin infliximab until the final 6 months of observation, the duration of follow-up evaluation was too short to determine how many patients required chronic infliximab therapy. However, a recent Danish retrospective analysis of children with CD treated with infliximab showed that only 29% experienced a prolonged clinical response after an induction regimen without subsequent infusions, and that another 42% required maintenance therapy.9 If long-term infliximab is ultimately necessary to keep children with CD free of corticosteroids, studies will be needed to determine the safety and efficacy of maintenance infliximab in this population.
Supplementary Data Note: To access the supplementary material accompanying this article, visit the online version of Clinical Gastroenterology and Hepatology at www.cghjournal.org. References 1. Markowitz J, Grancher K, Kohn N, et al. Immunomodulatory therapy for pediatric inflammatory bowel disease: changing patterns of use, 1990 –2000. Am J Gastroenterol 2002;97:928 –932. 2. Faubion WA Jr, Loftus EV Jr, Harmsen WS, et al. The natural history of corticosteroid therapy for inflammatory bowel disease: a population-based study. Gastroenterology 2001;121:255–260. 3. Markowitz J, Grancher K, Kohn N, et al. A multicenter trial of 6-mercaptopurine and prednisone in children with newly diagnosed Crohn’s disease. Gastroenterology 2000;119:895–902. 4. Hyams JS, Markowitz J, Wyllie R. Use of infliximab in the treatment of Crohn’s disease in children and adolescents. J Pediatr 2000; 137:192–196. 5. Lamireau T, Cezard JP, Dabadie A, et al. Efficacy and tolerance of infliximab in children and adolescents with Crohn’s disease. Inflamm Bowel Dis 2004;10:745–750. 6. Munkholm P, Langholz E, Davidsen M, et al. Frequency of glucocorticoid resistance and dependency in Crohn’s disease. Gut 1994; 35:360 –362. 7. Escher JC, Taminiau JA, Nieuwenhuis EE, et al. Treatment of inflammatory bowel disease in childhood: best available evidence. Inflamm Bowel Dis 2003;9:34 –58.
September 2006
8. Beattie RM. Therapy of Crohn’s disease in childhood. Paediatr Drugs 2000;2:193–203. 9. Wewer V, Riis S, Vind I, et al. Infliximab dependency in a national cohort of children with Crohn’s disease. J Pediatr Gastroenterol Nutr 2006;42:40 – 45.
Address requests for reprints to: James Markowitz, MD, Division of Pediatric Gastroenterology, Schneider Children’s Hospital, 269-01 76th Avenue, New Hyde Park, New York 11040. e-mail: jmarkowi@ nshs.edu; fax: (718) 962-2908. Supported by grants from Centocor, Inc. (Malvern, PA), AstraZeneca (Wilmington, DE), Reach Out for Youth With Ileitis and Colitis (Melville, NY), and the collaborating institutions. The operational expenses for the Registry Data Center are supported by grants from Centocor (Horsham, PA), AstraZeneca (Wilmington, DE), and Reach Out for Youth with Ileitis
CORTICOSTEROIDS IN PEDIATRIC CROHN’S DISEASE
1129
and Colitis (Melville, NY), and from grants from the participating centers. Vasundhara Tolia has received grants, honoraria, and/or consultancy fees from Centocor. Drs Markowitz, Hyams, and Rosh have received honoraria and/or consultancy fees from Prometheus Labs. James Markowitz has received research support from and has served as a Consultant to Centocor. Jeffrey Hyams has received research support from Centocor, and has served as a Consultant to Centocor, Abbott, and ELAN. Subra Kugathasan participated as a Consultant (advisory panel in designing clinical trials) for Centocor and Abbott. Joel Rosh has received research grant support from Proctor and Gamble. The authors are deeply indebted to the following research coordinators whose efforts greatly facilitated the performance of this study: Patricia Davis, Kathy Grancher, Valerie Grant, Annette Langseder, Ruth Singleton, Anna Zholudev, George Kay, Gail Waltz, Kim Boyer, Shari Huffman, Cathy Williams, Rebecca Abood, Rosemary Nagy, Carol Rudman, Myrna Miller, Vivian Abadom, Janet Trotta, and Laura Defaveri.